Unusual binding properties of the SH3 domain of the yeast actin-binding protein Abp1: structural and functional analysis

Abp1p is an actin-binding protein that plays a central role in the organization of Saccharomyces cerevisiae actin cytoskeleton. By a combination of two-hybrid and phage-display approaches, we have identified six new ligands of the Abp1-SH3 domain. None of these SH3-mediated novel interactions was detected in recent all genome high throughput protein interaction projects. Here we show that the SH3-mediated association of Abp1p with the Ser/Thr kinases Prk1p and Ark1p is essential for their localization to actin cortical patches. The Abp1-SH3 domain has a rather unusual binding specificity, because its target peptides contain the tetrapentapeptide +XXXPXXPX+PXXL with positive charges flanking the polyproline core on both sides. Here we present the structure of the Abp1-SH3 domain solved at 1.3-A resolution. The peptide-binding pockets in the SH3 domain are flanked by two acidic residues that are uncommon at those positions in the SH3 domain family. We have shown by site-directed mutagenesis that one of these negatively charged side chains may be the key determinant for the preference for non-classical ligands.     

Evidence for physical and functional interactions among two Saccharomyces cerevisiae SH3 domain proteins, an adenylyl cyclase-associated protein and the actin cytoskeleton.

In a variety of organisms, a number of proteins associated with the cortical actin cytoskeleton contain SH3 domains, suggesting that these domains may provide the physical basis for functional interactions among structural and regulatory proteins in the actin cytoskeleton. We present evidence that SH3 domains mediate at least two independent functions of the Saccharomyces cerevisiae actin-binding protein Abp1p in vivo. Abp1p contains a single SH3 domain that has recently been shown to bind in vitro to the adenylyl cyclase-associated protein Srv2p. Immunofluorescence analysis of Srv2p subcellular localization in strains carrying mutations in either ABP1 or SRV2 reveals that the Abp1p SH3 domain mediates the normal association of Srv2p with the cortical actin cytoskeleton. We also show that a site in Abp1p itself is specifically bound by the SH3 domain of the actin-associated protein Rvs167p. Genetic analysis provides evidence that Abp1p and Rvs167p have functions that are closely interrelated. Abp1 null mutations, like rvs167 mutations, result in defects in sporulation and reduced viability under certain suboptimal growth conditions. In addition, mutations in ABP1 and RVS167 yield similar profiles of genetic "synthetic lethal" interactions when combined with mutations in genes encoding other cytoskeletal components. Mutations which specifically disrupt the SH3 domain-mediated interaction between Abp1p and Srv2p, however, show none of the shared phenotypes of abp1 and rvs167 mutations. We conclude that the Abp1p SH3 domain mediates the association of Srv2p with the cortical actin cytoskeleton, and that Abp1p performs a distinct function that is likely to involve binding by the Rvs167p SH3 domain. Overall, work presented here illustrates how SH3 domains can integrate the activities of multiple actin cytoskeleton proteins in response to varying environmental conditions.     

A conserved proline-rich region of the Saccharomyces cerevisiae cyclase-associated protein binds SH3 domains and modulates cytoskeletal localization.

Saccharomyces cerevisiae cyclase-associated protein (CAP or Srv2p) is multifunctional. The N-terminal third of CAP binds to adenylyl cyclase and has been implicated in adenylyl cyclase activation in vivo. The widely conserved C-terminal domain of CAP binds to monomeric actin and serves an important cytoskeletal regulatory function in vivo. In addition, all CAP homologs contain a centrally located proline-rich region which has no previously identified function. Recently, SH3 (Src homology 3) domains were shown to bind to proline-rich regions of proteins. Here we report that the proline-rich region of CAP is recognized by the SH3 domains of several proteins, including the yeast actin-associated protein Abp1p. Immunolocalization experiments demonstrate that CAP colocalizes with cortical actin-containing structures in vivo and that a region of CAP containing the SH3 domain binding site is required for this localization. We also demonstrate that the SH3 domain of yeast Abp1p and that of the yeast RAS protein guanine nucleotide exchange factor Cdc25p complex with adenylyl cyclase in vitro. Interestingly, the binding of the Cdc25p SH3 domain is not mediated by CAP and therefore may involve direct binding to adenylyl cyclase or to an unidentified protein which complexes with adenylyl cyclase. We also found that CAP homologous from Schizosaccharomyces pombe and humans bind SH3 domains. The human protein binds most strongly to the SH3 domain from the abl proto-oncogene. These observations identify CAP as an SH3 domain-binding protein and suggest that CAP mediates interactions between SH3 domain proteins and monomeric actin.     

Protein-protein interaction affinity plays a crucial role in controlling the Sho1p-mediated signal transduction pathway in yeast

Protein-protein interactions are required for most cellular functions, yet little is known about the relationship between protein-protein interaction affinity and biological activity. To investigate this issue, we engineered a series of mutants that incrementally reduced the affinity of the yeast Sho1p SH3 domain for its in vivo target, the MAP kinase kinase Pbs2p. We demonstrate a strong linear correlation between the binding energy of these mutants and quantitative in vivo outputs from the HOG high-osmolarity response pathway controlled by Sho1p. In addition, we find that reduction in binding affinity for the correct target within this pathway causes a proportional increase in misactivation of the related mating pheromone response pathway and that strong binding affinity alone does not guarantee efficient biological activity. Our experiments also indicate that a second binding surface on the Sho1p SH3 domain is required for its proper in vivo function.     

Mammalian Abp1, a signal-responsive F-actin-binding protein, links the actin cytoskeleton to endocytosis via the GTPase dynamin

The actin cytoskeleton has been implicated in endocytosis, yet few molecular links to the endocytic machinery have been established. Here we show that the mammalian F-actin-binding protein Abp1 (SH3P7/HIP-55) can functionally link the actin cytoskeleton to dynamin, a GTPase that functions in endocytosis. Abp1 binds directly to dynamin in vitro through its SH3 domain. Coimmunoprecipitation and colocalization studies demonstrated the in vivo relevance of this interaction. In neurons, mammalian Abp1 and dynamin colocalized at actin-rich sites proximal to the cell body during synaptogenesis. In fibroblasts, mAbp1 appeared at dynamin-rich sites of endocytosis upon growth factor stimulation. To test whether Abp1 functions in endocytosis, we overexpressed several Abp1 constructs in Cos-7 cells and assayed receptor-mediated endocytosis. While overexpression of Abp1's actin-binding modules did not interfere with endocytosis, overexpression of the SH3 domain led to a potent block of transferrin uptake. This implicates the Abp1/dynamin interaction in endocytic function. The endocytosis block was rescued by cooverexpression of dynamin. Since the addition of the actin-binding modules of Abp1 to the SH3 domain construct also fully restored endocytosis, Abp1 may support endocytosis by combining its SH3 domain interactions with cytoskeletal functions in response to signaling cascades converging on this linker protein.     

The importance of conserved features of yeast actin-binding protein 1 (abp1p): the conditional nature of essentiality

Saccharomyces cerevisiae Actin-Binding Protein 1 (Abp1p) is a member of the Abp1 family of proteins, which are in diverse organisms including fungi, nematodes, flies, and mammals. All proteins in this family possess an N-terminal Actin Depolymerizing Factor Homology (ADF-H) domain, a central Proline-Rich Region (PRR), and a C-terminal SH3 domain. In this study, we employed sequence analysis to identify additional conserved features of the family, including sequences rich in proline, glutamic acid, serine, and threonine amino acids (PEST), which are found in all family members examined, and two motifs, Conserved Fungal Motifs 1 and 2 (CFM1 and CFM2), that are conserved in fungi. We also discovered that, similar to its mammalian homologs, Abp1p is phosphorylated in its PRR. This phosphorylation is mediated by the Cdc28p and Pho85p kinases, and it protects Abp1p from proteolysis mediated by the conserved PEST sequences. We provide evidence for an intramolecular interaction between the PRR region and SH3 domain that may be affected by phosphorylation. Although deletion of CFM1 alone caused no detectable phenotype in any genetic backgrounds or conditions tested, deletion of this motif resulted in a significant reduction of growth when it was combined with a deletion of the ADF-H domain. Importantly, this result demonstrates that deletion of highly conserved domains on its own may produce no phenotype unless the domains are assayed in conjunction with deletions of other functionally important elements within the same protein. Detection of this type of intragenic synthetic lethality provides an important approach for understanding the function of individual protein domains or motifs.     

A synthetic lethal screen identifies a role for the cortical actin patch/endocytosis complex in the response to nutrient deprivation in Saccharomyces cerevisiae

Saccharomyces cerevisiae whi2Delta cells are unable to halt cell division in response to nutrient limitation and are sensitive to a wide variety of stresses. A synthetic lethal screen resulted in the isolation of siw mutants that had a phenotype similar to that of whi2Delta. Among these were mutations affecting SIW14, FEN2, SLT2, and THR4. Fluid-phase endocytosis is severely reduced or abolished in whi2Delta, siw14Delta, fen2Delta, and thr4Delta mutants. Furthermore, whi2Delta and siw14Delta mutants produce large actin clumps in stationary phase similar to those seen in prk1Delta ark1Delta mutants defective in protein kinases that regulate the actin cytoskeleton. Overexpression of SIW14 in a prk1Delta strain resulted in a loss of cortical actin patches and cables and was lethal. Overexpression of SIW14 also rescued the caffeine sensitivity of the slt2 mutant isolated in the screen, but this was not due to alteration of the phosphorylation state of Slt2. These observations suggest that endocytosis and the organization of the actin cytoskeleton are required for the proper response to nutrient limitation. This hypothesis is supported by the observation that rvs161Delta, sla1Delta, sla2Delta, vrp1Delta, ypt51Delta, ypt52Delta, and end3Delta mutations, which disrupt the organization of the actin cytoskeleton and/or reduce endocytosis, have a phenotype similar to that of whi2Delta mutants.     

A Conserved residue in the yeast Bem1p SH3 domain maintains the high level of binding specificity required for function

The yeast Bem1p SH3b and Nbp2p SH3 domains are unusual because they bind to peptides containing the same consensus sequence, yet they perform different functions and display low sequence similarity. In this work, by analyzing the interactions of these domains with six biologically relevant peptides containing the consensus sequence, they are shown to possess finely tuned and distinct binding specificities. We also identify a residue in the Bem1p SH3b domain that inhibits binding, yet is highly conserved for the purpose of preventing nonspecific interactions. Substitution of this residue results in a marked reduction of in vivo function that is caused by titration of the domain away from its proper targets through nonspecific interactions with other proteins. This work provides a clear illustration of the importance of intrinsic binding specificity for the function of protein-protein interaction modules, and the key role of "negative" interactions in determining the specificity of a domain.     

Sla1p is a functionally modular component of the yeast cortical actin cytoskeleton required for correct localization of both Rho1p-GTPase and Sla2p, a protein with talin homology

SLA1 was identified previously in budding yeast in a genetic screen for mutations that caused a requirement for the actin-binding protein Abp1p and was shown to be required for normal cortical actin patch structure and organization. Here, we show that Sla1p, like Abp1p, localizes to cortical actin patches. Furthermore, Sla1p is required for the correct localization of Sla2p, an actin-binding protein with homology to talin implicated in endocytosis, and the Rho1p-GTPase, which is associated with the cell wall biosynthesis enzyme beta-1,3-glucan synthase. Mislocalization of Rho1p in sla1 null cells is consistent with our observation that these cells possess aberrantly thick cell walls. Expression of mutant forms of Sla1p in which specific domains were deleted showed that the phenotypes associated with the full deletion are functionally separable. In particular, a region of Sla1p encompassing the third SH3 domain is important for growth at high temperatures, for the organization of cortical actin patches, and for nucleated actin assembly in a permeabilized yeast cell assay. The apparent redundancy between Sla1p and Abp1p resides in the C-terminal repeat region of Sla1p. A homologue of SLA1 was identified in Schizosaccharomyces pombe. Despite relatively low overall sequence homology, this gene was able to rescue the temperature sensitivity associated with a deletion of SLA1 in Saccharomyces cerevisiae.     

Activation of the Arp2/3 complex by the actin filament binding protein Abp1p

The actin-related protein (Arp) 2/3 complex plays a central role in assembly of actin networks. Because distinct actin-based structures mediate diverse processes, many proteins are likely to make spatially and temporally regulated interactions with the Arp2/3 complex. We have isolated a new activator, Abp1p, which associates tightly with the yeast Arp2/3 complex. Abp1p contains two acidic sequences (DDW) similar to those found in SCAR/WASp proteins. We demonstrate that mutation of these sequences abolishes Arp2/3 complex activation in vitro. Genetic studies indicate that this activity is important for Abp1p functions in vivo. In contrast to SCAR/WASp proteins, Abp1p binds specifically to actin filaments, not monomers. Actin filament binding is mediated by the ADF/cofilin homology (ADF-H) domain of Abp1p and is required for Arp2/3 complex activation in vitro. We demonstrate that Abp1p recruits Arp2/3 complex to the sides of filaments, suggesting a novel mechanism of activation. Studies in yeast and mammalian cells indicate that Abp1p is involved functionally in endocytosis. Based on these results, we speculate that Abp1p may link Arp2/3-mediated actin assembly to a specific step in endocytosis.     

Phosphoregulation of Arp2/3-dependent actin assembly during receptor-mediated endocytosis

In both yeast and mammals, endocytic internalization is accompanied by a transient burst of actin polymerization. The yeast protein kinases Prk1p and Ark1p, which are related to the mammalian proteins GAK and AAK1, are key regulators of this process. However, the molecular mechanism(s) by which they regulate actin assembly at endocytic sites have not yet been determined. The Eps15-like yeast protein Pan1p is a Prk1p substrate that is essential for endocytic internalization and for proper actin organization. Pan1p is an Arp2/3 activator and here we show that this activity is dependent on F-actin binding. Mutation of all 15 Prk1p-targeted threonines in Pan1p to alanines mimicked the ark1Delta prk1Delta phenotype, demonstrating that Pan1p is a key Prk1p target in vivo. Moreover, phosphorylation by Prk1p inhibited the ability of Pan1p to bind to F-actin and to activate the Arp2/3 complex, thereby identifying the endocytic phosphoregulation mechanism of Prk1p. We conclude that Prk1p phosphorylation of Pan1p shuts off Arp2/3-mediated actin polymerization on endocytic vesicles, allowing them to fuse with endosomes.     

The design of a hyperstable mutant of the Abp1p SH3 domain by sequence alignment analysis

We have characterized the thermodynamic stability of the SH3 domain from the Saccharomyces cerevisiae Abp1p protein and found it to be relatively low compared to most other SH3 domains, with a Tm of 60 degrees C and a deltaGu of 3.08 kcal/mol. Analysis of a large alignment of SH3 domains led to the identification of atypical residues at eight positions in the wild-type Abp1p SH3 domain sequence that were subsequently replaced by the residue seen most frequently at that position in the alignment. Three of the eight mutants constructed in this way displayed increases in Tm ranging from 8 to 15 degrees C with concomitant increases in deltaGu of up to 1.4 kcal/mol. The effects of these substitutions on folding thermodynamics and kinetics were entirely additive, and a mutant containing all three was dramatically stabilized with a Tm greater than 90 degrees C and a deltaGu more than double that of the wild-type domain. The folding rate of this hyperstable mutant was 10-fold faster than wild-type, while its unfolding rate was fivefold slower. All of the stabilized mutants were still able to bind a target peptide with wild-type affinity. We have analyzed the stabilizing amino acid substitutions isolated in this study and several other similar sequence alignment based studies. In approximately 25% of cases, increased stability can be explained by enhanced propensity of the substituted residue for the local backbone conformation at the mutagenized site.     

Functional redundancies, distinct localizations and interactions among three fission yeast homologs of centromere protein-B

Several members of protein families that are conserved in higher eukaryotes are known to play a role in centromere function in the fission yeast Schizosaccharomyces pombe, including two homologs of the mammalian centromere protein CENP-B, Abp1p and Cbh1p. Here we characterize a third S. pombe CENP-B homolog, Cbh2p (CENP-B homolog 2). cbh2Delta strains exhibited a modest elevation in minichromosome loss, similar to cbh1Delta or abp1Delta strains. cbh2Delta cbh1Delta strains showed little difference in growth or minichromosome loss rate when compared to single deletion strains. In contrast, cbh2Delta abp1Delta strains displayed dramatic morphological and chromosome segregation defects, as well as enhancement of the slow-growth phenotype of abp1Delta strains, indicating partial functional redundancy between these proteins. Both cbh2Delta abp1Delta and cbh1Delta abp1Delta strains also showed strongly enhanced sensitivity to a microtubule-destabilizing drug, consistent with a mitotic function for these proteins. Cbh2p was localized to the central core and core-associated repeat regions of centromeric heterochromatin, but not at several other centromeric and arm locations tested. Thus, like its mammalian counterpart, Cbh2p appeared to be localized exclusively to a portion of centromeric heterochromatin. In contrast, Abp1p was detected in both centromeric heterochromatin and in chromatin at two of three replication origins tested. Cbh2p and Abp1p homodimerized in the budding yeast two-hybrid assay, but did not interact with each other. These results suggest that indirect cooperation between different CENP-B-like DNA binding proteins with partially overlapping chromatin distributions helps to establish a functional centromere.     

Structural and functional dissection of the Abp1 ADFH actin-binding domain reveals versatile in vivo adapter functions

Abp1 is a multidomain protein that regulates the Arp2/3 complex and links proteins involved in endocytosis to the actin cytoskeleton. All of the proposed cellular functions of Abp1 involve actin filament binding, yet the actin binding site(s) on Abp1 have not been identified, nor has the importance of actin binding for Abp1 localization and function in vivo been tested. Here, we report the crystal structure of the Saccharomyces cerevisiae Abp1 actin-binding actin depolymerizing factor homology (ADFH) domain and dissect its activities by mutagenesis. Abp1-ADFH domain and ADF/cofilin structures are similar, and they use conserved surfaces to bind actin; however, there are also key differences that help explain their differential effects on actin dynamics. Using point mutations, we demonstrate that actin binding is required for localization of Abp1 in vivo, the lethality caused by Abp1 overexpression, and the ability of Abp1 to activate Arp2/3 complex. Furthermore, we genetically uncouple ABP1 functions that overlap with SAC6, SLA1, and SLA2, showing they require distinct combinations of activities and interactions. Together, our data provide the first structural and functional view of the Abp1-actin interaction and show that Abp1 has distinct cellular roles as an adapter, linking different sets of ligands for each function.     

The phosphoinositide phosphatase Sjl2 is recruited to cortical actin patches in the control of vesicle formation and fission during endocytosis

The Saccharomyces cerevisiae synaptojanin-like proteins (Sjl1, Sjl2, and Sjl3) are phosphoinositide (PI) phosphatases that regulate PI metabolism in the control of actin organization and membrane trafficking. However, the primary sites of action for each of the yeast synaptojanin-like proteins remain unclear. In this study, we show that Sjl2 is localized to cortical actin patches, sites of endocytosis. Cortical recruitment of Sjl2 requires the actin patch component Abp1. Consistent with this, the SH3 domain-containing protein Abp1 physically associates with Sjl2 through its proline-rich domain. Furthermore, abp1Delta mutations confer defects resembling loss of SJL2; sjl1Delta abp1Delta double-mutant cells exhibit invaginated plasma membranes and impaired endocytosis, findings similar to those for sjl1Delta sjl2Delta mutant cells. Thus, Abp1 acts as an adaptor protein in the localization or concentration of Sjl2 during late stages of endocytic vesicle formation. Overexpression of the Hip1-related protein Sla2 delayed the formation of extended plasma membrane invaginations in sjl2ts cells, indicating that Sla2 may become limiting or misregulated in cells with impaired PI phosphatase activity. Consistent with this, the cortical actin patch protein Sla2 is mislocalized in sjl1Delta sjl2Delta mutant cells. Together, our studies suggest that PI metabolism by the synaptojanin-like proteins coordinately directs actin dynamics and membrane invagination, in part by regulation of Sla2.     

Binding of cytoplasmic proteins to the CD19 intracellular domain is high affinity, competitive, and multimeric

CD19 is required for the development of B1 and marginal zone B cells, for Ab responses, and for B cell memory. CD19 immunoprecipitates contain a complex of cytoplasmic proteins, including Lyn, Vav, phospholipase Cgamma2 (PLCgamma2), Grb2, and the p85 subunit of phosphatidylinositol 3-kinase. Which of these bind directly to CD19 and the strengths of the interactions are unknown. These issues are important in understanding the signaling functions of CD19, which are crucial for normal B cell physiology. Using purified, recombinant proteins, we now show that each of these signaling proteins contains at least one Src homology 2 (SH2) domain that interacts directly with the phosphorylated CD19 cytoplasmic domain. The affinities of binding of the SH2 domains of Vav, p85, and Grb2 to CD19 are each in the nanomolar range by surface plasmon resonance (Biacore) analysis. Binding of Lyn and PLCgamma2 do not fit 1:1 modeling. However, analyses of binding data (Lyn) and competition experiments (PLCgamma2) suggest that these bind with comparable affinity. Competition experiments demonstrate that SH2 domains whose binding is dependent on the same CD19 tyrosine(s) compete for binding, but these SH2 domains do not impede binding of different SH2 domains to other CD19 tyrosines. We conclude that binding to the CD19 cytoplasmic domain is multimeric, high affinity, and competitive. The high affinity of the interactions also suggests that tyrosines that were nonessential in vivo are nevertheless functional. A preliminary structural model suggests that CD19 forms a signaling complex containing multiple cytoplasmic proteins in close proximity to each other and to the plasma membrane.     

In vivo analysis of the domains of yeast Rvs167p suggests Rvs167p function is mediated through multiple protein interactions.

Morphological changes during cell division in the yeast Saccharomyces cerevisiae are controlled by cell-cycle regulators. The Pcl-Pho85p kinase complex has been implicated in the regulation of the actin cytoskeleton at least in part through Rvs167p. Rvs167p consists of three domains called BAR, GPA, and SH3. Using a two-hybrid assay, we demonstrated that each region of Rvs167p participates in protein-protein interactions: the BAR domain bound the BAR domain of another Rvs167p protein and that of Rvs161p, the GPA region bound Pcl2p, and the SH3 domain bound Abp1p. We identified Rvs167p as a Las17p/Bee1p-interacting protein in a two-hybrid screen and showed that Las17p/Bee1p bound the SH3 domain of Rvs167p. We tested the extent to which the Rvs167p protein domains rescued phenotypes associated with deletion of RVS167: salt sensitivity, random budding, and endocytosis and sporulation defects. The BAR domain was sufficient for full or partial rescue of all rvs167 mutant phenotypes tested but not required for the sporulation defect for which the SH3 domain was also sufficient. Overexpression of Rvs167p inhibits cell growth. The BAR domain was essential for this inhibition and the SH3 domain had only a minor effect. Rvs167p may link the cell cycle regulator Pcl-Pho85p kinase and the actin cytoskeleton. We propose that Rvs167p is activated by phosphorylation in its GPA region by the Pcl-Pho85p kinase. Upon activation, Rvs167p enters a multiprotein complex, making critical contacts in its BAR domain and redundant or minor contacts with its SH3 domain.     

Comparative genomics and disorder prediction identify biologically relevant SH3 protein interactions

Protein interaction networks are an important part of the post-genomic effort to integrate a part-list view of the cell into system-level understanding. Using a set of 11 yeast genomes we show that combining comparative genomics and secondary structure information greatly increases consensus-based prediction of SH3 targets. Benchmarking of our method against positive and negative standards gave 83% accuracy with 26% coverage. The concept of an optimal divergence time for effective comparative genomics studies was analyzed, demonstrating that genomes of species that diverged very recently from Saccharomyces cerevisiae (S. mikatae, S. bayanus, and S. paradoxus), or a long time ago (Neurospora crassa and Schizosaccharomyces pombe), contain less information for accurate prediction of SH3 targets than species within the optimal divergence time proposed. We also show here that intrinsically disordered SH3 domain targets are more probable sites of interaction than equivalent sites within ordered regions. Our findings highlight several novel S. cerevisiae SH3 protein interactions, the value of selection of optimal divergence times in comparative genomics studies, and the importance of intrinsic disorder for protein interactions. Based on our results we propose novel roles for the S. cerevisiae proteins Abp1p in endocytosis and Hse1p in endosome protein sorting.     

The SH3 domain of the S. cerevisiae Cdc25p binds adenylyl cyclase and facilitates Ras regulation of cAMP signalling

Cdc25 and Ras are two proteins required for cAMP signalling in the budding yeast Saccharomyces cerevisiae. Cdc25 is the guanine nucleotide exchange protein that activates Ras. Ras, in turn, activates adenylyl cyclase. Cdc25 has a Src homology 3 (SH3) domain near the N-terminus and a catalytic domain in the C-terminal region. We find that a point mutation in the SH3 domain attenuates cAMP signalling in response to glucose feeding. Furthermore, we demonstrate, by using recombinant adenylyl cyclase and Cdc25, that the SH3 domain of Cdc25 can bind directly to adenylyl cyclase. Binding was specific, because the SH3 domain of Abp1p (actin-binding protein 1), which binds the 70,000 Mr subunit of adenylyl cyclase, CAP/Srv2, failed to bind adenylyl cyclase. A binding site for Cdc25-SH3 localised to the C-terminal catalytic region of adenylyl cyclase. Finally, pre-incubation with Ras enhanced the SH3-bound adenylyl cyclase activity. These studies suggest that a direct interaction between Cdc25 and adenylyl cyclase promotes efficient assembly of the adenylyl cyclase complex.